← All comparisons

Gemma 4 E4B (Non-reasoning) vs Claude 4.5 Haiku (Reasoning)

Google vs Anthropic — side-by-side benchmark comparison

Gemma 4 E4B (Non-reasoning)Claude 4.5 Haiku (Reasoning)
Intelligence Index14.837.1
Coding Index6.432.6
Math Index83.7
Output speed (tok/s)0.0142.2
Blended price ($/1M)$0.00$2.19
Time to first token (s)0.00s10.48s
aime
aime 2583.7%
artificial analysis coding index6.4032.60
artificial analysis intelligence index14.8037.10
artificial analysis math index83.70
gpqa54.9%67.2%
hle4.7%9.7%
ifbench40.5%54.3%
lcr18.0%70.3%
livecodebench61.5%
math 500
mmlu pro76.0%
scicode3.9%43.3%
tau226.0%54.7%
terminalbench hard7.6%27.3%

Benchmark data from Artificial Analysis.